How do we understand seemingly inconsistent behavior in foreign policy? This is the question Eleonora Mattiacci takes on in Volatile States in International Politics. In International Relations (IR) scholarship, the focus tends to be on consistent change : either towards conflict in terms of escalation or towards cooperation in terms of reconciliation. Seldom do scholars endeavor to explain inconsistency in state action.
H-Diplo | Robert Jervis International Security Studies Forum
Roundtable Review 16-2
Eleonora Mattiacci. Volatile States in International Politics. Oxford University Press, 2023. ISBN: 9780197638682.
13 September 2024 |PDF: https://issforum.org/to/jrt16-2 | Website: rjissf.org | X @HDiplo
Editor: Diane Labrosse
Commissioning Editor: Danielle Lupton
Production Editor: Christopher Ball
Pre-Production Copy Editor: Bethany Keenan
Contents
Introduction by Susan Hannah Allen, University of Mississippi 2
Review by Jeff Carter, Appalachian State University. 5
Review by Lisa Langdon Koch, Claremont McKenna College. 9
Review by Roseanne W. McManus, Pennsylvania State University. 12
Response by Eleonora Mattiacci, Amherst College. 16
Introduction by Susan Hannah Allen, University of Mississippi
How do we understand seemingly inconsistent behavior in foreign policy? This is the question Eleonora Mattiacci takes on in Volatile States in International Politics. In International Relations (IR) scholarship, the focus tends to be on consistent change : either towards conflict in terms of escalation or towards cooperation in terms of reconciliation. Seldom do scholars endeavor to explain inconsistency in state action.
When we observe inconsistency, Mattiacci suggests that foreign policy volatility is driven by the intersection of two distinct processes—competition among domestic groups and a state’s growth in relative power. When states have more power, we know they have more policy options, but state action is also more variable when divergent domestic political interests are competing to influence foreign policy. To test this central idea, Mattiacci employs both quantitative large-N analysis and qualitative case study exploration, examining volatility among international rivals in the 20th century with a special focus on US foreign policy towards France during the 1950s and 1960s.
The three reviews in this roundtable reflect the broad potential for applying Mattiacci’s insights. Each of the three reviewers finds something of value as they use the lens of variability to think about their own areas of research while also highlighting places where Mattiacci could do more or does not go far enough with her assessment of volatility.
In his essay, while Jeff Carter considers the benefits of clarifying volatility, both in terms of conceptual clarity and creating a valid measure for concept, he challenges the importance that Mattiacci’s theory places on relative power, suggesting that absolute power might be what is moving the needle. He suggests that states with more power in absolute terms have a wider range of foreign policy options, which increases the potential for volatility. That potential is not contingent on the power of a rival. He also raises questions about the consequences of volatility. The theory suggests that increased volatility and uncertainty should affect foreign policy decision-making, but the empirical testing does not explore the outcomes of these decision-making processes.
Lisa Koch also discusses the role that power plays in Mattiacci’s book. She begins with praise for the innovative measures that Mattiacci develops to capture volatility among rivals. On the other hand, she is more critical of the case selection in the large-N portion of the analysis. She argues that by focusing on rivals, Mattiacci’s test only encompasses countries that have engaged in significant conflict. These countries are militarily strong and may be more likely to resort to military action. Koch does note that Mattiacci helps to counterbalance this limitation by focusing on a pair of allies (the US and France between 1954 and 1966) in the case-study chapter. Drawing on her own work, Koch notes how nuclear acquisition by France influenced the relationship between the two countries and suggests that including nuclear proliferation in the earlier measure of power or as an additional variable for power might be useful.
Roseanne McManus brings a psychological perspective to her review, noting that humans want to identify patterns. This helps to explain why IR scholars have primarily focused on consistent state behavior rather than volatile behavior. McManus, however, notes that there is one area of research where unpredictable behavior has been front and center: the Madman Theory. Taking the discussion of volatility down from the state level and considering how it might be connected to individual leader psychology pushes Mattiacci’s theory in a new direction. Rather than assuming that only negative effects stem from volatility, the Madman Theory opens up the potential for positive impacts.
In her response, Eleonora Mattiacci reflects on the nature of power in the international system. In defense of the importance of relative power to her theory, she notes that the absolute power of the United States has not changed much in the twenty-first century, but drawing on Koch’s comments regarding nuclear weapons, she argues that as countries like Iran and North Korea have pursued proliferation, US policy towards those countries has increased in variability. In addition, she seconds McManus’s suggestion that additional case studies of volatile policies could improve our understanding of the domestic interests.
Volatile behavior is by definition unpredictable. This challenges scholars to think about the models and explanations that we use for state behaviors. We want to focus on stable patterns, which are easier to predict for scholars and are perceived as safer for the inhabitants of planet Earth. Mattiacci’s book is a good reminder that because state behavior can be volatile, we must create dynamic theories that can encompass that reality. She hopes that this book will serve as a call for greater interest from scholars in the variation around that mean behavior rather than the tradition focus on the average behavior of states. All of the contributors included in this forum do a nice job beginning a conversation about how our models and theories can stretch and adjust to accommodate the realities of the unpredictable world we live in.
Contributors:
Eleonora Mattiacci is an Assistant Professor of Political Science at Amherst College, where she runs the International Relations Lab. She has published on the causes of conflict, with a specific focus on civil wars, nuclear weapons, and public diplomacy. She is the author of Volatile States in International Politics (Oxford University Press, 2023). Her current research explores environmental destruction during wars.
Susan Hannah Allen is Professor of Political Science at the University of Mississippi. Her research interests include air power, economic sanctions, and consequences of coercion. She recently published Bargaining and the UN Security Council: Setting the Global Agenda with Oxford University Press and she has published in multiple journals including International Studies Quarterly, Journal of Conflict Resolution, and Journal of Global Security Studies.
Jeff Carter is an Associate Professor of Political Science at Appalachian State University and the Co-Director of the Correlates of War Project. His research focuses on the relationship between political leaders and interstate conflict, how governments pay for war, and state making processes. His work has been published in the American Political Science Review, Journal of Politics, International Studies Quarterly, and other outlets.
Lisa Langdon Koch is an Assistant Professor of Government at Claremont McKenna College. She is the author of Nuclear Decisions: Changing the Course of Nuclear Weapons Programs (Oxford University Press, 2023).
Roseanne McManus is an Associate Professor at the Pennsylvania State University. Her research focuses on international security and particularly how states can signal their intentions in conflict bargaining. She is the author of two articles on the Madman Theory, published in the British Journal of Political Science and Security Studies. Her research has also appeared in the American Political Science Review, Journal of Politics, International Organization, and other journals. Her book, Statements of Resolve: Achieving Coercive Credibility in International Conflict, was published by from Cambridge University Press in 2017.
Review by Jeff Carter, Appalachian State University
It is standard for scholars, politicians, and pundits to make claims about the changing and unpredictable nature of the world. Such pronouncements are rarely backed up with anything beyond an anecdote or two, much less a coherent explanation for why the changes are occurring. Volatile States in International Politics by Eleonora Mattiacci makes a strong case that states’ foreign policies are not only dynamic, but also sometimes experience periods of volatility where they swing inconsistently between cooperative and conflictual policies. Mattiacci has written an excellent book on a topic that is poorly understood. The book provides both an intuitive framework for understanding volatility and empirical analyses that demonstrate states’ material capabilities and domestic politics interact to influence the volatility of their foreign policies.
Volatile States consists of five chapters. Following an introduction that explains the importance of studying volatility, chapter 2 presents the book’s theoretical framework. Chapter 3 describes how Mattiacci measures foreign policy volatility, while chapter 4 presents the book’s statistical analyses of volatility among 36 pairs of international rivals between 1948 and 1992. Chapter 5 uses the book’s theoretical framework to inform a qualitative analysis of volatility in French foreign policy towards the US in the 1950s and 1960s. The book concludes with a discussion of its implications for our understanding of volatility and suggestions for how scholars might extend its insights to analyze volatility’s influence on a number of other topics central to International Relations.
The book makes three substantial contributions to the systematic study of world politics. The first is that it provides conceptual clarity about what volatility is and, just as importantly, what volatility is not. Volatile foreign policy behavior is “behavior that shifts towards more cooperation and more conflict in a way that looks inconsistent to observers” (3). Volatility, then, is inconsistent change in the nature of a state’s foreign policies. The fundamentally inconsistent nature of volatile foreign policy makes it hard for states to predict one another’s behavior and differentiates volatility from foreign policy that is characterized by systematic cycles, periods of escalation and de-escalation, and instability. The explicit focus on volatility sets Mattiacci’s work apart from other work on foreign policy and international relations dynamics, including, but not limited to, Kurt Gaubatz on cycles; Michael Colaresi, William Thompson, and Karen Rasler on escalation and de-escalation; and Paul Diehl and Gary Goertz on instability.[1]
The book’s second major contribution is its theoretical framework for explaining foreign policy volatility. Mattiacci argues that increases in a state’s relative power and the presence of domestic groups with heterogenous political interests are associated with more volatile foreign policy behavior. The core ideas are that 1) a state’s capabilities influence the range of foreign policies it can pursue; 2) heterogeneity in the preferences held by groups that can compete for control of a state’s foreign policy influences the extent to which the state implements diverse and varied foreign policies; and 3) the joint presence of sufficient capabilities to implement a wide range of policies and sufficiently heterogeneous domestic interests with potential access to power makes volatile foreign policy more likely. Taking a step back, what Mattiacci has done is cleverly combine insights from research on work on foreign policy substitutability by Glenn Palmer and Cliff Morgan[2] and others and the immense literature on the influence of domestic politics on decision-making[3] to explain foreign policy volatility within the context of the classic “opportunity and willingness” paradigm popularized by Benjamin Most and Harvey Starr: capabilities determine a state’s opportunity to implement a range of cooperative and conflictual foreign policies while heterogeneous domestic interests determine its willingness to do so.[4] The result is an intuitive model of foreign policy that is novel but feels familiar.
I think the book’s third significant contribution is its strategy for measuring volatility. Constructing valid and consistent indicators of latent concepts across space and time is more difficult than most scholars realize. Without minimizing other issues, I think that the most substantial challenge to generating measures of volatility is that an indicator needs to be able to distinguish between foreign policy behavior that reflects volatility, and foreign policy behavior that reflects cycles, escalatory and de-escalatory dynamics, or instability. This is especially tricky because a combination of cycles, escalatory and de-escalatory dynamics, and/or instability can make a state appear to inconsistently change between cooperative and conflictual foreign policies when this is not the case. Failing to account for this would result in indicators that systematically overstate foreign policy volatility and, consequently, lead to incorrect inferences.
Mattiacci’s strategy for constructing measures that account for this issue is clever. It is also relatively technical. Readers who are interested in the methodological details should consult the book’s appendix. In non-technical terms, Mattiacci uses time-series techniques to isolate deviations in a state’s systematic patterns of foreign policy (i.e., cycles, escalation and de-escalation, and [in]stability) towards a rival. This produces a variable that identifies inconsistent changes in a state’s cooperative and conflictual foreign policies over time. This indicator of foreign policy volatility, and the measurement strategy more generally, reflects the careful work of a thoughtful and talented methodologist.
The book’s conceptual clarity, theoretical framework, and measurement strategy regarding an often overlooked and misunderstood topic represent real contributions to the study of foreign policy and International Relations. That said, in two cases the book missed the mark and an opportunity to make a larger contribution, respectively.
The first is that it is not clear why a state’s potential foreign policy volatility is a function of its relative power and not its absolute power. Mattiacci argues that relative power serves as the “permissive condition for volatile behavior” (24) and explicitly clarifies that it is a state’s “relative, as opposed to absolute, power” (26) that shapes the range of available foreign policies.
I’m skeptical. To see why, it is worth taking a step back and asking why power matters for volatility. According to Mattiacci, “power increases the policies available to a country because it makes more options available to them, both cooperative and conflictual” (16) and “the mechanism at play here is resource availability (25). These are claims that a state’s absolute pool of resources influences the range of foreign policies it can pursue.
I think a small thought experiment can help clarify the issue. Consider two scenarios. In scenario one, an exogenous shock results in State A and State B having twice as many resources at time t+1 than they had at time t. Is the range of foreign policies State A and State B can implement at time t+1 greater than the range of foreign policies they could implement at time t? If the answer is yes, this is consistent with absolute power influencing potential volatility. If the answer is no, this is consistent with relative power influencing potential volatility, but it is inconsistent with the claim that resource availability underlies potential volatility.
Now consider a second scenario. In scenario two, an exogenous shock doubles State A’s resources but triples State B’s resources. If resource availability is the mechanism that underlies potential volatility, then the increase in resources should allow both State A and State B to implement a wider range of policies. However, if potential volatility is a function of relative capabilities, a shock that doubles State A’s resources reduces the range of policies it can implement because the same shock tripled State B’s resources. If this is the case, then resource availability cannot be the mechanism underlying potential foreign policy volatility.
Beyond the thought experiment, it is absolute and not relative resources that drives the ability of a state to implement more cooperative and conflictual foreign policies in Palmer and Morgan’s model of foreign policy substitutability[5] that the book’s theoretical framework builds on. Relative power will clearly shape the probability of success for various foreign policies. Within the opportunity and willingness paradigm, though, the probability of success influences the attractiveness of a policy and, consequently, a state’s relative willingness to implement it, not its ability or opportunity to do so. If resource availability is the mechanism underlying what policies a state can implement, as Mattiacci claims, a state’s absolute power represents the “permissive condition” for volatility.
My second critique is that the book would have benefited from an analysis of how a country’s foreign policy volatility influences subsequent foreign policy and/or international outcomes. Readers are told throughout the book, but especially in the introductory and concluding chapters, that volatility is important in no small part because of its likely negative consequences for world politics. This is certainly plausible given the increased uncertainty associated with volatility and the effects of uncertainty on other actors’ decision-making. It would have been nice to see a systematic analysis of, for example, how one state’s foreign policy volatility influences the likelihood of its rival initiating an interstate dispute against it. The fact that there is no dedicated chapter to a consequence of volatility does not take anything away from the book so much as it represents a missed opportunity to make an additional contribution.
Eleonora Mattiacci has written an important book on a poorly understood topic. It should shape how scholars think about foreign policy dynamics and spur future work on how volatility influences on world politics. I highly recommend it to anyone who wants to understand foreign policy and international relations.
Review by Lisa Langdon Koch, Claremont McKenna College
In Volatile States in International Politics, Eleonora Mattiacci challenges readers to think beyond what she describes as “consistent forms of change” (3). That phrase may sound like a contradiction in terms, but it describes exactly what Mattiacci urges us to realize: that we tend to want to examine phenomena that fit an expected pattern, like periods of escalating conflict between states, or periods of deepening cooperation. Instead, Mattiacci argues that we should seek to understand the portion of state behavior that is volatile: behavior that switches inconsistently between cooperation and conflict. While volatile behavior usually sparks fear—by definition, volatility is unpredictable and unstable—that fear may be misplaced. After all, unpredictable bursts of cooperation, however temporary, would be preferable to a stable pattern of brutal violence. Stability may not be normatively good, and volatility may not be normatively bad.
Mattiacci’s novel concept helps us make sense of behavior that seems chaotic. It is a welcome addition to the rivalries literature, building most directly on Michael Colaresi, Karen Rasler, and William Thompson’s Strategic Rivalries in World Politics. Colaresi, Rasler, and Thompson reconceptualized rivals as acting within in a “stream of conflict,” in which both cooperative and conflictual behavior are more likely to be perceived as threatening.[6] Mattiacci now directs us to look explicitly at rivals’ actions in the liminal spaces to understand whether rivals’ patterns of behavior are volatile or consistent. By looking beyond conflict in rivalries, and observing the non-conflictual behavior as well, we can better understand the paths that states choose to take.
Consider US behavior in the international system during and after the Trump administration. Was the United States during the Trump years a volatile state, swinging erratically between conflict and cooperation with foreign states? That seems to have been the case in the context of the US-North Korea rivalry, in which sudden and surprising changes between cooperation and conflict were “whiplash inducing.”[7] Or was it instead more of a norm-breaking actor, exhibiting behavior that was unusual but that consistently moved toward conflict? That may have been more true of the US-China rivalry, in which an early trend toward cooperation transitioned to an overall trend toward conflict.[8] Indeed, would the United States be considered volatile under the first two years of the Biden administration, as some policies abruptly reverted to a pre-Trump state, while others, such as trade relations with China, remained fairly steady? Examining the full range of one state’s behavior toward another will generate insights that focusing only on the conflict would obscure.
Why, then, would a state behave in a volatile way? Mattiacci argues that volatility emerges from a combination of two processes. First, a state must be increasing in military power relative to its rival. Irrespective of whether the state is weaker or stronger than its rival, the more relative power the state gains, the more the range of that state’s policy options increases. Then, if that condition has been met, there must be competition among domestic groups with diverging foreign policy interests. Mattiacci uses societal cleavages to identify this second condition. The increase in the range of options must be present in order for competing domestic interests to generate the volatile behavior that swings between and among those options.
The statistical techniques Mattiacci uses to conduct quantitative tests of volatility among rivals are innovative and clearly described. In particular, I was eager to learn how to actually measure volatility, and I was not disappointed. I will not be able to do the measure justice in this short review; readers should dive into chapter 3 for the step-by-step approach. Suffice it to say that within a pairing of states, one state’s inconsistent behavioral shifts are captured by measuring how far each observed act of cooperation or conflict is from that state’s average behavior. The shifts are plotted and presented visually (figs. 3.1-3.5). On face, the plots support the foundational argument that volatile change is different from change that follows a pattern. Alongside the concept of volatile behavior in international politics, Mattiacci’s operationalization of volatile behavior ranks, in my mind, as the second major contribution of the book.
While there is much to like in this book, there are a few issues to consider further. The careful and thorough large-N analysis presented in chapter 4 examines highly conflictual rivals and provides evidence in support of the application of this theory to that subsample of states. Because the book is theoretically centered around volatility within rivals’ relationships, it makes the most sense to test the theory using the subset of rivals. However, because the large-N analysis is confined only to the thirty-six pairs of rivals that have been prone to the highest levels of conflict, it is difficult to evaluate the strength of the theory more broadly.
Additionally, within the subset of rival states, selecting only high-conflict pairs may bias the dataset toward actors that are militarily strong enough to engage in significant conflict behavior, and/or that are more likely to select coercive strategies.[9] For example, the substantively largest, significant finding (table 4.3) is that volatility in country A’s rival begets volatility in country A. This may apply across all rivalries, or it could be a particularly strong explanator in high-conflict pairings. Of course, difficult tradeoffs must be made when deciding how to empirically test a theory. An attempt to collect fine-grained event history data on a very large grouping of states, such as the full subpopulation of rivals and allies, would surely be a colossal undertaking, and would be unsuited to a single book project.
The book’s case study, France and the US between 1954 and 1966, departs from the large-N study, as the case is drawn from a separate subpopulation: not rivals, but allies. The selected time span includes both the notoriously unstable Fourth Republic and the first years of the Fifth Republic under the strong leadership of President Charles de Gaulle, and a wide range of state behavior is presented, in keeping with the breadth of the theory. Interestingly, French military strength in relation to US military strength, which is measured by dividing the French national material capabilities score by the US score in each year, ranges from 0.152 to 0.115, varying only by 0.04 during 1954-1966 (figure 5.1).[10] Mattiacci rightly points out that France’s nuclear weapons acquisition served as an important factor in French leaders’ beliefs about what options France would have in pursuing its interests (98). Rather than relying only on the military capability ratio, which does not include nuclear weapons capabilities, as the measure of relative power, both the statistical analysis and case analysis would likely benefit from a variable for nuclear weapons possession. This would also align nicely with many of the nuclear weapons-based examples of volatility that Mattiacci discusses throughout the book, from China in 1964, to India and Pakistan post-1998, to North Korea post-2006.
Overall, Mattiacci’s engaging work succeeds in making the case that volatility is a phenomenon deserving of our attention. Volatile States will inspire many readers to reexamine their assumptions about stability and instability in state behavior. Mattiacci’s concept of volatility as a distinctive feature of interstate relationships is an important contribution that offers scholars a new perspective from which to examine what we observe in the world.
Review by Roseanne W. McManus, Pennsylvania State University
Psychology tells us that individuals are constantly looking to identify patterns and overarching narratives in the world around them. This tendency is so strong that we sometimes think we have identified a pattern in a series of events even when there is none. Scholars are not necessarily immune from this. As Eleanora Mattiacci points out in Volatile States in International Politics, scholarship in International Relations has tended to focus on analyzing behavior by states that exhibits a clear pattern or narrative, such as spiraling conflict or increasing cooperation.[11] Yet some states, or pairs of states, exhibit behavior that does not fit this mold. Instead, they veer from cooperation to conflict in an unpredictable manner. Mattiacci’s important and insightful book takes on the unique task of explaining this behavior.
Mattiacci defines volatility as “behavior that shifts toward more cooperation and more conflict in a way that looks inconsistent to observers” (3). Her introductory chapter contrasts this with other patterns of state behavior, explaining how volatility is different from related behavior patterns, such as cyclical or unstable behavior.
Mattiacci offers an interactive theory to explain the causes of volatile state behavior. The root cause that she identifies is a situation in which domestic groups are deeply divided over foreign policy goals, and more than one group has the ability to influence foreign policy. Intuitively, just like two drivers fighting for control of the steering wheel will cause a car to swerve, different groups pursuing different goals whenever they are able to influence foreign policy will result in a volatile foreign policy. However, Mattiacci says this dynamic alone cannot fully explain volatile behavior because some states are limited in their volatility due to a lack of options. Only states with sufficient power on the international stage have enough conflictual and cooperative foreign policy options available to allow volatile behavior to clearly manifest. Therefore, Mattiacci predicts that volatile behavior will be produced by a combination of military power and divisions among domestic groups with influence on foreign policy.
Mattiacci’s empirical analysis has three components. First, she produces a novel measure of
volatility using event data and the residuals of a Box-Jenkins procedure (56-58).[12] Second, she presents regressions predicting states’ levels of volatility based on a standard measure of relative military power and a novel index measuring the extent to which heterogeneous interests within states can influence foreign policy. This analysis finds support for her theory. Third, she presents a qualitative case study of volatility in France’s relations with the United States.
This book addresses an important and unique topic, overcoming a previously unrecognized bias in the literature toward studying more consistent behavior. It offers clear argumentation and a sophisticated use of time-series analysis, combined with qualitative evidence to illustrate the causal logic. The findings have important implications for theories of both mistrust and cooperation.
As with all research, this book has some weaknesses. The specific mechanisms through which the divergent preferences of domestic groups aggregate to influence foreign policy could be theorized in more depth. This shortcoming of the theory influences the empirical design too, as the index of the ability of competing domestic interests to influence foreign policy is measured roughly, accounting for few nuances in domestic politics (79). Of course, this book is the first take on a highly novel topic, so it is understandable that not every nuance is deeply explored. Future research will be able to build on this work and consider more nuances.
Mattiacci primarily situates her book in contrast with the literature on cooperation and conflict. Although this is a valid and useful juxtaposition, I would like to draw some new connections by focusing the remainder of my review on the relationship with a different literature; namely, literature on the Madman Theory. The Madman Theory, originally articulated by Daniel Ellsberg, holds that being viewed as a madman provides advantages in conflict bargaining. Recent research by Avidit Acharya and Edoardo Grillo; James Boys; Nicholas Campbell-Seremetis; Roseanne McManus; Joshua Schwartz; Samuel Seitz and Caitlin Talmadge; and Seanon Wong has further explored this argument and added some nuances.[13] While the definition of a madman differs among these works, many include volatility as a core component. For example, Ellsberg identifies one type of madness as being “at least 1% likely to do anything” (emphasis original).[14]
Mattiacci’s work and the Madman Theory offer different conceptions of the source of volatility. Mattiacci focuses on competition for control of foreign policy. Ultimately, hers is a rationalist explanation in which individuals who act rationally based on their own incentives produce an outcome that only looks volatile from the outside. As Mattiacci emphasizes in her conclusion, the volatility produced by intra-government competition is actually somewhat predictable (126). Having a solid understanding of the domestic politics of a country can help to predict its volatility, if not the precise swings in policy.
In contrast, the Madman Theory focuses on a single leader, with full or at least dominant control of the government, who acts volatilely. Under the Madman Theory, the source of volatility may either be the leader’s psychology (in the case of genuine madmen) or a strategic attempt to keep opponents off balance (in the case of leaders who feign madness). Returning to an earlier analogy, volatility in the Madman Theory comes not from struggle for control of the steering wheel, but from a single erratic driver. This might suggest the potential for volatility to go to greater extremes because there is no competing domestic force acting to push the country back the other way when it goes too far in one direction.
Mattiacci attempts to rule out leader-focused explanations for state volatility in her study by including control variables for leader gender and combat experience, but neither of these is a perfect proxy for leader psychology. Thus, more work remains to be done to compare and investigate the explanatory power of these different potential sources of volatility. Moreover, these are not necessarily the only potential sources of volatility. It could be possible for a leadership group to act volatilely even in the absence of competition between the members. This could result from a shared ideology that values volatility or from inconsistent decision-making procedures that produce different outcomes even when preferences are the same.
Mattiacci’s book and the Madman Theory also differ in the effects of volatility that they consider. Mattiacci considers various negative effects, including spirals of fear and distrust. However, she concludes that these negative effects can be ameliorated if the sources of volatility are properly understood. With proper understanding, observers will interpret a volatile state as reacting to different domestic constituencies rather than deliberately behaving in an untrustworthy manner.
Rather than viewing volatility as a negative quality that can possibly be ameliorated, the Madman Theory views it more positively, arguing that it can bring advantages in international conflict bargaining. The fundamental reason for this difference is that the Madman Theory is focused on conveying resolve, whereas Mattiacci’s work is more focused on establishing trust. It is common for theories about resolve and trust to speak past each other, but it is also unfortunate because the very actions that convey resolve may also promote distrust.[15] Some recent interpretations of the Madman Theory have recognized this,[16] and Mattiacci’s emphasize on how volatility can cause distrust speaks to these interpretations. However, a solution to the distrust caused by volatility may be more elusive if observers believe it is caused by the unstable psychology of a leader rather than the aggregation of different preferences held by domestic actors.
In sum, Mattiacci’s book offers a major innovation by recognizing, conceptualizing, and theorizing about the previously underexplored concept of volatility. It has important connections to the Madman Theory and also to broader literatures on distrust and cooperation. It is my hope that this book will spur even more research on the causes and effects of volatility in International Relations.
Response by Eleonora Mattiacci, Amherst College
My colleagues Jeff Carter, Lisa Langdon Koch, and Roseanne McManus have reviewed my book with great perceptiveness and care, and Danielle Lupton has done a masterful job spearheading this exchange, as perhaps only an expert on leaders and resolve could do.[17] I am very grateful to them all.
Volatile States in International Politics laments that International Relations scholars have been far too shy about addressing volatility in the international arena (15-16). Reviews on this roundtable reverse this state of affairs. In my colleagues’ expert hands, volatility works as a kaleidoscope of sorts. It rearranges our images of the international arena to reveal novel phenomena to explore.
Koch, for example, builds on her research on the dangers of bundling together different countries’ experiences when studying proliferation.[18] She invites us to explore proliferators’ reactions to the entering and exiting of phases of volatile behavior by the US. McManus puts volatility in conversation with her own research on Madman Theory.[19] She ponders how scholars’ explanations of the international system change depending on whether they interpret unpredictability as showing resolve or as fomenting mistrust. Carter muses on the effects of volatility by connecting it to his own studies of conflict initiation.[20] He urges us to think about how being exposed to volatile behavior for prolonged periods of time might make states more willing to initiate disputes.
Leveraging the study of volatility to enrich our understanding of international politics, these reviewers generously honor my book’s goals. To boot, their criticisms are thoughtful and productive. They point to areas for improvement in the book and for interesting directions in the exploration of volatility. In particular, they do so in four areas.
Some of the reviews wrestle with the role that power plays in my explanation for volatility. Rising relative capabilities increase the probability of volatility (25-29), acting as a permissive condition. Koch’s suggestion to include a measure of nuclear weapons possession when assessing the role of power is smart. One could even broaden the scope of the analysis. We could also research whether those countries that are one step removed from acquiring nuclear weapons (the so-called nuclear latent states) become more prone to volatility.[21] Comparing the effects of acquisition and latency on volatility can teach us a lot about the effects of the issue of technological capability thresholds.
Apropos of power, Carter is right when he claims that “a state’s absolute pool of resources influences the range of foreign policy it can pursue.” But I am not convinced that the book should have used a measure of absolute power to capture the effect of power on volatility. Koch’s reflection on volatility and nuclear weapons states might provide some inspiration for making this point more concrete. US absolute power did not change much during the 2000s. But, as Koch suggests, during the process of nuclear-weapons acquisition that culminated in the 2006 nuclear explosion, North Korea’s power relative to the US has gradually increased. As a consequence, US power relative to North Korea has decreased. With that decrease also comes a decrease in the options at the disposal of US leaders to deal with North Korea. Because these resources offer a permissive condition for volatility, the probability of volatile US behavior toward North Korea also has decreased. Focusing on relative power reveals such important nuances on the effects of power on volatility.
McManus asks whether the measure of domestic interests that I employ is fine-tuned enough to measure all the different nuances of domestic interests. There is a tradeoff between precision and applicability. The measure needed to be nimble enough to reliably capture domestic forces in (very) different institutional and interest configurations. An encompassing, broad test in turn is important for a book that seeks to show how pervasive volatility is in international politics. The book uses an additive index that addresses three straightforward questions. “First, can different domestic groups access power? […] Second, if groups can alternate, do they represent a heterogeneous set of domestic preferences? […]Third, is the electoral system] proportional?” (79). Chapter 5 complements this approach by using historical statistics and archival sources to retrace how the domestic context (interests and institutions) interacted with power to increase volatility in French behavior toward the United States. But McManus’s point is an important one, especially for furthering the study of volatility: it is paramount to further clarify the mechanisms connecting domestic interests and volatility. More case studies might be of great value here.
These reviews ask important questions about the theory-testing section of the book. Koch asks whether focusing the large-N analysis on rivals can really reassure us that my argument can also explain volatility in non-rivalrous countries. She is right that focusing on rivals poses clear tradeoffs in terms of internal/external validity. But the pros still outweigh the cons. The biggest advantage has to do with the quality of the data. Event data, which I use to retrace volatility, rely on newswires. For this reason, “event data are better suited to capture episodes of conflict and cooperation that occur between states that are often in the news” (75). Rivals often appear in newswires reports, because they repeatedly engage in conflict.[22] Interactions between countries that are not as often in the news are less likely to be covered. So it is harder to accurately retrace their foreign policy behaviors in search for volatility. But accurately retracing behavior is very important to measure volatility. Even more, because rivalries are such well-studied interactions, scholars may think they know all there is to know about them. Showing that rivalries vary in terms of how volatile they are can present big payoffs as the book seeks to establish the importance of understanding a less well studied phenomenon such as volatility.
Still, Koch’s point is well taken: if we had better data, would we want to engage in an analysis of volatility among other pairs of states? And if we did, what would we find? Expanding the test to new countries will certainly be a fruitful way to find out more about volatility. Nothing in the theory I propose prevents power and interests from spurring volatility between countries that are neither allies nor rivals. And such expansion might be paramount to clarifying many of the current debates in international politics, including those on reputation, credible commitments, trust, and audience costs (125-134). Some of the reviews suggest further explorations of the effects of volatility. I was hoping my book would whet IR scholars’ appetite for these kinds of questions.
Volatility, the book argues, challenges trust and ignites conflict (12-13,123-127). Understanding the diffusion of volatile state behavior can shed light on important system-level emerging properties, such as the frequency of interactions, the emergence of norms and expectations, and polarity (133-134). The book uses primary documents to reveal the degree to which US policy-makers grew suspicious of their French allies after facing years of volatile French behavior over the period 1954–1963(118-119). Americans learned not to trust the French. They came to see them as “political competitors pursuing divergent goals” (119). Such primary sources are crucial to support claims made in the book regarding the deleterious consequences of volatile behavior in the relations between states. They “uniquely reveal how perceptions of volatility impacted decision makers’ thinking and behavior, both publicly and covertly” (93).
But Carter’s point stands: exploring “how one state’s foreign policy volatility influences the likelihood of its rival initiating an interstate dispute against it” would have enriched the book. Building on Carter’s own research, it would be interesting to explore whether dovish or hawkish leaders react differently to observing volatility and if such different reactions shape their propensity for conflict.[23]
McManus instead asks whether volatility be something other than “a negative quality that can possibly ameliorated.” Volatile States posits that it can be. When we understand that volatility reveals competition among domestic groups, volatile behavior becomes an important source of information regarding states’ preferences. As such, it can help build stronger relationships between countries (125-129). But I agree that it would be interesting to further explore cases where volatile countries rip off benefits from their unpredictability, instead of (or in addition to) compromising trust and igniting conflict. For example, do volatile countries extract concessions from their opponents at times? [24]
I was delighted to read McManus’s smart reflection on Madman Theory and volatility. As Carter notes, Volatile States is in conversation with theories of foreign policy and international relations dynamics. McManus is correct that adding a connection to Madman Theory might be fruitful. McManus is of course also right when she claims that volatility can be part of a “madman” strategy. But “madman” strategies do not need be volatile. In perhaps the quintessential Madman Theory case, President Nixon tried to signal that he was a “madman” by making it clear that he would feel no remorse when escalating violence during 1969. To project the image of a “madman,” he did not act in a volatile manner. Instead, he consistently embraced more conflict and more violence.[25]
Another difference between the two theoretical approaches is one of focus. Madman Theory zooms in on the role of individuals. Due perhaps to its Cold War roots, Madman Theory has mostly been investigated in the American and Soviet institutional contexts.[26] Both contexts tend to provide ample room to maneuver for leaders in foreign policy. As such, the geographical and historical focus of the theory raise the question of whether Madmen Theory tends to over-estimate the role of individuals in foreign policy. Volatile States tries instead to offer measurement tools that capture volatility in multiple, diverse institutional contexts.
I agree with McManus, however: volatility in international politics should be put in conversation with Madman Theory. McManus’s own research is perhaps uniquely positioned to open a dialogue between these two areas of study. This is the case because McManus broadens the empirical and theoretical scope of Madman Theory by delving into what earns leaders the label of “madman.”[27] So, do observers tend to perceive those labeled as “madman” types as being more volatile than they are? And does volatile behavior tend to precede or follow a “madman” reputation? These questions are sure to enrich both studies of Madman Theory and of volatility.
To (mis)quote an often (ab)used maxim: change is the only constant in international politics.[28] But International Relations scholars are fellow human beings. They tend to seek patterns and focus on change that is consistent (18-20). And so most of the change we learn about in IR looks like conflict spirals or reoccurring cycles of increased cooperation and conflict. Volatile States in International Politics provides methodological and theoretical tools for IR scholars to study inconsistent, unpredictable change. It then describes how doing so “can contribute to ameliorating volatility’s worst consequences; […] can provide the missing piece to numerous IR puzzles; and […] can illuminate fertile grounds for further investigations of volatility in other areas of international politics, including systemic politics and nonstate actors’ behaviors” (139).
My colleagues’ thoughtful, engaging reviews delineate what my book achieved, where it fell short, and how to move forward. I could hardly think of a better starting point for what will hopefully be a steady increase in our investigations of volatile behaviors in international politics.
[1] Kurt Taylor Gaubatz, “Election Cycles and War,” Journal of Conflict Resolution 35:2 (1991): 212-244; Michael P. Colaresi, Karen Rasler, and William R. Thompson, Strategic Rivalries in World Politics: Position, Space and Conflict Escalation (Cambridge, Cambridge University Press, 2008); Paul Diehl and Gary Goertz, War and Peace in International Rivalry (Ann Arbor: University of Michigan Press, 2001).
[2] Glenn Palmer and T. Clifton Morgan, A Theory of Foreign Policy (Princeton: Princeton University Press, 2006).
[3] While many works could be cited here, research by Ashley Leeds and Michaela Mattes on changes in domestic sources of leaders’ support on foreign policy is closely related to Mattiacci’s argument. Among others, Brett Ashley Leeds and Michaela Mattes, Domestic Interests, Democracy, and Foreign Policy Change (Cambridge: Cambridge University Press, 2022).
[4] Benjamin A. Most and Harvey Starr, “Conceptualizing ‘War’: Consequences for Theory and Research,” Journal of Conflict Resolution 27:1 (1983): 137-159.
[5] Palmer and Morgan, A Theory of Foreign Policy.
[6] Michael P. Colaresi, Karen Rasler, and William R. Thompson, Strategic Rivalries in World Politics (Cambridge: Cambridge University Press, 2007).
[7] Victor Cha and Katrin Fraser Katz, “The Right Way to Coerce North Korea: Ending the Threat Without Going to War,” Foreign Affairs 97:3 (2018): 87-100, 87.
[8] Klaus Larres, “Trump’s Trade Wars: America, China, Europe, and Global Disorder,” Journal of Transatlantic Studies 18:1 (March 1, 2020): 103-29.
[9] Colaresi, Rasler, and Thompson, Strategic Rivalries in World Politics, 52-55.
[10] J. David Singer, “Reconstructing the Correlates of War Dataset on Material Capabilities of States, 1816–1985,” International Interactions 14:2 (1987): 115-132.
[11] See, for example, Robert Jervis, Perception and Misperception in International Politics (Princeton: Princeton University Press, 1976); Andrew H. Kydd, Trust and Mistrust in International Relations (Princeton: Princeton University Press, 2005).
[12] A Box-Jenkins procedure compares different model specifications to identify the one that best represents the data’s temporal dynamics.
[13] Daniel Ellsberg, “The Political Uses of Madness,” Lecture given at the Lowell Institute of the Boston Public Library, 26 March 1959; https://ia800102.us.archive.org/20/items/ThePoliticalUsesOfMadness/ELS005-001.pdf; Avidit Acharya and Edoardo Grillo, “War with Crazy Types.” Political Science Research and Methods 3:2 (2015): 281-307; James D. Boys, “The Unpredictability Factor: Nixon, Trump and the Application of the Madman Theory in US Grand Strategy,” Cambridge Review of International Affairs 34:3 (2021): 430-451; Nicholas Campbell-Seremetis, “Fanatics, Fools, and Madmen: How Perceptions of Bias, Incompetence, and Irrationality Impact International Politics,” PhD diss. (University of Chicago, 2020); Roseanne W. McManus,”Revisiting the Madman Theory: Evaluating the Impact of Different Forms of Perceived Madness in Coercive Bargaining,” Security Studies 28:5 (2019): 976-1009; McManus, “Crazy like a Fox? Are Leaders with Reputations for Madness More Successful at International Coercion?” British Journal of Political Science 51:1 (2021): 275-293; Joshua A. Schwartz, “Madman or Mad Genius? The International Benefits and Domestic Costs of the Madman Strategy.” Security Studies 32:2 (2022): 271-305; Samuel Seitz and Caitlin Talmadge, “The Predictable Hazards of Unpredictability: Why Madman Behavior Doesn’t Work.” Washington Quarterly 43:3 (2020): 31-46; Seanon Wong, “Stoics and Hotheads: Leaders’ Temperament, Anger, and the Expression of Resolve in Face-to-Face Diplomacy,” Journal of Global Security Studies 4:2 (2019): 190-208.
[14] Ellsberg, “The Political Uses of Madness,” 5.
[15] Andrew H. Kydd and McManus, “Threats and Assurances in Crisis Bargaining,” Journal of Conflict Resolution 61:2 (2017): 325-348.
[16] McManus, “Revisiting the Madman Theory: Evaluating the Impact of Different Forms of Perceived Madness in Coercive Bargaining.”
[17] Danielle L. Lupton, Reputation for Resolve: How Leaders Signal Determination in International Politics (Ithaca, NY: Cornell University Press, 2020).
[18] Lisa Langdon Koch, Nuclear Decisions: Changing the Course of Nuclear Weapons Programs (New York: Oxford University Press, 2023).
[19] Roseanne W. McManus, “Crazy Like a Fox? Are Leaders with Reputations for Madness More Successful at International Coercion?” British Journal of Political Science 51:1 (2021): 275-293.
[20] Jeff Carter and Charles E. Smith. “A Framework for Measuring Leaders’ Willingness to Use Force.” American Political Science Review 114:4 (2020): 1352-1358. Jeff Carter, “Leader Hawkishness, Political Survival, and Interstate Crises.” The Journal of Politics 86:1 (2024): 337-51.
[21] Benjamin Jones, Eleonora Mattiacci, and Timothy Nordstrom, “How Leader’s Type Shapes the Effect of Nuclear Latency on Dispute Involvement.” Conflict Management and Peace Science 41:2 (2024): 177-193; Eleonora Mattiacci, Rupal N. Mehta, and Rachel Elizabeth Whitlark, “Atomic Ambiguity: Event Data Evidence on Nuclear Latency and International Cooperation.” Journal of Conflict Resolution 66:2 (2022): 272-296.
[22] Philip A. Schrodt, Erin M. Simpson, and Deborah J. Gerner, “Monitoring Conflict Using Automated Coding of Newswire Reports: a Comparison of Five Geographical Regions,” presented at the PRIO / Uppsala University / DECRG High-Level Scientific conference on Identifying Wars: Systematic Conflict Research and Its Utility in Conflict Resolution and Prevention, Uppsala, Sweden, 8-9 June 2001.
[23] Jeff Carter and Charles E. Smith, “A Framework for Measuring Leaders’ Willingness to Use Force,” American Political Science Review 114:4 (2020): 1352-1358; Jeff Carter, “Leader Hawkishness, Political Survival, and Interstate Crises,” The Journal of Politics 86:1 (2024): 337-351.
[24] See also Robert Schub, “Volatile States in International Politics. By Eleonora Mattiacci. New York: Oxford University Press, 2023,” Perspectives on Politics, 2024, DOI: https://doi.org/10.1017/S1537592723002748..
[25] Scott D. Sagan, and Jeremi Suri, “The Madman Nuclear Alert: Secrecy, Signaling, and Safety in October 1969,” International Security 27:4 (2003): 150-183.
[26] McManus’s research is a rare and refreshing exception. Roseanne W. McManus, “Revisiting the Madman Theory: Evaluating the Impact of Different Forms of Perceived Madness in Coercive Bargaining,” Security Studies 28:5 (2019): 976-1009.
[27] McManus, “Crazy Like a Fox? Are Leaders with Reputations for Madness More Successful at International Coercion?” British Journal of Political Science 51:1 (2021): 275-293.
[28] Keith Begley, “The only constant is… misunderstanding of Heraclitus,” Edinburgh University Press Blog. Available here: https://euppublishingblog.com/2021/07/19/misunderstanding-of-heraclitus/.